Re: [PATCH net] selftests: drv-net: devmem: Skip when hugetlbfs isn't supported

[Maxime Chevallier]

On 10/27/26 22:15, Jakub Kicinski wrote:
> On Mon, 26 Oct 2026 10:38:50 +0100 Maxime Chevallier wrote:
>> On 10/23/26 19:52, Jakub Kicinski wrote:
>>> On Wed, 21 Oct 2026 11:08:51 +0200 Maxime Chevallier wrote:  
>>>> The devmem selftest requires accessing /proc/sys/vm/nr_hugepages,
>>>> which is only available when hugetlbfs is supported. This isn't the
>>>> case on all CPU architectures, for example it isn't supported on ARM
>>>> 32-bits on cores that don't support LPAE.
>>>>
>>>> On these devices, the devmem kselftest doesn't skip gracefully  
>>>
>>> We don't require graceful skipping. If anything IMO hiding the stack
>>> traces isn't great.
>>>
>>> If the check is extraneous and makes us fail some tests which don't
>>> actually require huge pages please make that clear in the commit
>>> message. Otherwise the net benefit of the changes is unclear.  
>>
>> hugetlbfs isn't required for devmem AFAICS, but the blamed commit
>> moved the access to nr_hugepages before the actual "is devmem
>> actually supported" test.
>>
>> I don't expect the full test suite to gracefully run/skip for devmem
>> on these arm32 platforms TBH, it's likely that devmem is only something
>> that's been used on platforms that support hugetlbfs anyways, but
>> strictly speaking both features are unrelated.
> 
> For max clarity I should have simply asked - does it make any test which
> previously failed pass on any of your platforms? (incl. some WIP
> patches that'd add devmem support) That's basically the criteria 
> I think we should use.

I see, then we can drop that patch. It would go from a "file not found"
python exception trace to a skip. This is a nice criteria for me to follow,
I'll move on to other tests :)

Maxime